Personalizing health and food advices by semantic enrichment of multilingual cross-domain questions

Web search engines help in retrieving the scattered information from the Web, albeit with a number of limitations. They can't understand or enrich the user's natural language questions easily or offer the recommendation that fits the user's exact needs. Health and food information are examples of critical domains where the users have a lot of questions that need to be understood well, enriched and processed to retrieve answers that match the user's needs. Using the personalization and the semantic techniques help us to propose a framework that enriches the user's questions and retrieves more relevant results. In this paper, we analyze the user's preferences related to the health and food domains, and then propose a user's profile ontology that represents these preferences and map them to the pre-defined domain ontologies. The proposed framework has been implemented and the experimental results show promising results with user satisfaction.